Skip to content

January 2, 2011

Why You Should Be A/B Testing Your Website

A/B Testing TipsA/B testing may seem like a complicated term but it is actually quite simple. A definition of A/B testing is the following:

Using A/B testing, at its most basic, randomly shows visitors one version of a page – version (A) or the alternative, version (B) – and tracks any change to behavior based on the version they saw.

The reason why this is so important can best be expressed with an example. You create an ecommerce website where the call to action to purchase the item on the Home Page is a green circular button that says “Purchase”. The green circle was arbitrarily chosen and your conversion rate is 10%.

Question: What if that arbitrarily chosen green circular button was a blue circular button instead? Would the conversion rate increase/decrease?

Although it might not appear to make a difference, the impact that A/B testing can have on your conversion rate can be substantial. Perhaps by changing the color of the button to blue your conversion rate increases by 10%. Possibly by changing the shape to a square the conversion rate will again increase. Through A/B testing, websites can increase their conversion rate by 2X to 10X. That is a 200% to 1000% increase.

A/B testing is also quite easy to implement with Google Optimizer. However, a recent Y Combinator backed startup called Optimizely (http://www.optimizely.com) is looking to make this process even simpler by allowing you to change the design without even having to touch the code. You can read more about the company on their website.

If you are interested in building a successful internet company, regardless of whether the business is ecommerce based, A/B testing is something you should implement. It is hard to find an initiative that will have such a high return on investment.

Here is more information on how A/B testing relates to paid advertising.

Get Adobe Flash player